Membrane distillation crystallization for brine mining and zero liquid discharge: opportunities, challenges, and recent progress

Youngkwon Choi a, Gayathri Naidu a, Long D. Nghiem a, Sangho Lee b and Saravanamuthu Vigneswaran *a
aFaculty of Engineering, University of Technology Sydney (UTS), P.O Box 123, Broadway, NSW 2007, Australia. E-mail:; Fax: +61 2 9514 2633; Tel: +61 2 9514 2641
bSchool of Civil and Environmental Engineering, Kookmin University, 77 Jeongneung-ro, Jeongneung-dong, Seongbuk-gu, Seoul 136-702, Republic of Korea

Received 22nd February 2019 , Accepted 7th May 2019

First published on 9th May 2019

Membrane distillation crystallization (MDC) is an emerging alternative for sustainable management of challenging hypersaline solutions such as seawater brine, produced water, and some industrial wastewaters. MDC combines two individual processes, specifically membrane distillation and crystallization to facilitate the extraction of clean water and separation of high purity salt contents from a brine solution. The potential of MDC for treating challenging hypersaline solutions has been suggested in numerous proof-of-concept studies and some lab-scale demonstration. Nonetheless, there still remain technical challenges which need to be studied before MDC can be economically implemented in practice. This review documents the basic concept of MDC and opportunities to strategically deploy this emerging technology platform for brine mining and zero liquid discharge (ZLD). This paper also discusses key technical challenges (including scaling prevention and membrane wetting which need to be addressed for successful field application) hindering the practical realization of MDC for full-scale operation as well as pathways to address these challenges. Different crystallization techniques such as reaction crystallization and drowning-out crystallization are covered here in order to offer suggestions on their suitability for MDC operation.

Water impact

Membrane distillation crystallization is a beneficial integrated treatment process that can simultaneously recover clean water and valuable resources in crystal forms. Its beneficial applications for various challenging water sources are discussed in this review. Challenges of the integrated process and factors that can improve its performance are evaluated in detail.

1. Introduction

The lack of potable water and the depletion of raw materials are two major issues that have emerged in recent decades due to the rapid increase in the world's population and improvements in standards of living.1 Membrane technology plays a major role in addressing the above issues. Specific additional treatments are required for improving water recovery strategies, operational costs, and quality of water produced, overcoming the impact of brine (i.e. concentrate) and determining how much of it is disposed of into the eco-system.2 Brine from seawater reverse osmosis (SWRO) plants and wastewater streams from oil, shale gas and coal seam gas production plants are some examples of challenging water treatment and discharge scenarios. These sources consist of high concentrations of various inorganic, organic and toxic components which can seriously endanger human and environmental health.

For this reason, it is essential to find and implement the best treatment and disposal methods.3–7 For example, in the SWRO process the water recovery factor ranges from 30 to 50% in seawater reverse osmosis plants, resulting in the generation of a huge amount of concentrate (brine).8,9 It is one of the biggest challenges in the SWRO desalination process. The extraction of oil and shale gas using hydraulic fracturing generates a huge amount of wastewater stream which is often known as produced water (PW). The major components of produced water are high concentrations of inorganic salt, organic compounds, heavy metals, oil, grease and production chemical compounds.7,10 The volume of produced water stream is nearly 70% of the total wastewater stream generated, and this amount is several times larger than the volume of recovered oil or gas.11

Membrane distillation (MD) is a promising option to treat brine and produced water.6,7,12–18 In the pressure-driven membrane process, the high concentration (high salt (inorganic) content in the solution) suppresses the mass transfer across the membrane. In the MD process, however, the effect of concentration on the mass transfer is not important because the driving force of MD is thermally induced by the vapor pressure difference between the hot feed stream and cold permeate stream. Thus, it allows MD to have a lower operating pressure. Moreover, a nonvolatile solution can be completely rejected because only vapor is transferred throughout the hydrophobic membrane, resulting in high quality/purity permeate water being produced.4,14,19 As a result, MD has much potential for the treatment of high concentration solutions.

Crystallization technology presents a good option for the treatment of high concentration and complex solutions, i.e. challenging solutions. Basically, the main idea behind crystallization technology is to control the saturation level utilizing specific methods such as reduction of temperature and removal of solvent. The crystallization process is one of the basic unit-operations for separation and purification of chemical matter in aqueous solution, and in fact, one of the most effective and powerful ways to produce valuable crystals from the parent solution. This is a solid–fluid separation process in which crystals are formed from the parent solution, and it has been employed in many industries such as the chemical, petrochemical, pharmaceutical and food sectors. The reason why this is the case is that crystallization exhibits wide-ranging applicability.20,21 More than 70% of pharmaceutical and chemical production processes involve crystallization.22 In particular, the batch cooling crystallization process is a widely used method for recovering valuable resources in the pharmaceutical industry.23,24 The crystallization technology has certain advantages such as a high recovery ratio of resources and recovery of high quality water.25 Although crystallization is widely used for purification and separation, the design and operation of a crystallizer still face many limitations.3,26 These include, for example, poor reproducibility in the final crystals' characteristics, limited supersaturation control and less ability to modulate the supersaturation generation rate.

The membrane distillation crystallization (MDC) process has the potential to solve the above issues. MD employed in the MDC process can remove/produce a solvent from an aqueous solution consisting of an inorganic solute, resulting in supersaturation. Consequently, the recovery of fresh water and valuable resources can be achieved at the same time. Due to the operational properties of both MD and crystallization, the integration/combination of MD with the crystallization process (MDC) has high potential for the treatment of high salinity solutions without additional post-treatment.9 MDC will result in an enhanced water recovery ratio, a reduced brine disposal problem, a faster crystallization rate and curtailment of crystals' induction time.27,28 MDC has been successfully applied for the recovery of fresh water and valuable resources from challenging solutions.4,5,8,9,11,29–33 The interesting aspect of this process is that: firstly, crystallization may require less energy compared to conventional evaporation crystallization; secondly, it produces crystalline materials from solution; and thirdly, clean water is recovered. Moreover, MD as part of MDC can improve the limited supersaturation control of crystallization due to continuous solvent removal.

This review reports and critically reviews the research conducted on the membrane distillation crystallization (MDC) process to date. The potential and feasibility of MDC in treatment of challenging solutions is discussed. This review also addresses operational problems such as crystal deposition/formation (scale) on the membrane, the wetting phenomenon and economic implications (energy consumption and operational costs). Other applicable crystallization techniques such as drowning-out crystallization and reaction crystallization technologies are noted here.

2. Crystallization/precipitation process

Crystallization/precipitation from solution is widely used by industries in their purification and solid–liquid separation processes.34,35 It is the basic unit-operation in the production of crystalline commodity products.36 The basis of crystallization is to achieve supersaturation of the solute in solution; when it is exceeded, crystallization and separation of crystals from the bulk solution occur. During the crystallization/precipitation process, it is important to control the solute's solubility so that the appropriate conditions to form crystals are facilitated. This is because the driving force of crystallization is the supersaturation of the solute in solution. Factors such as decline in solution temperature, solvent removal, and the chemical reaction between the solutes significantly affect the solubility of the solute in solution, leading to higher crystallization. The following phenomena make this possible: cooling crystallization, evaporation crystallization, reaction crystallization and drowning-out crystallization, respectively.25,35

The solubility of the solute is a function of the solution temperature; the influence of temperature on solubility may be positive, negative, or neutral. Fig. 1 shows the variation of the solubility tendency as a function of temperature. Positive temperature–solubility occurs when solubility increases as temperature increases (solute A (e.g. Na2SO4, MgCl2 and KCl)), while negative temperature–solubility occurs when solubility decreases as temperature increases (solute C (e.g. Ca(OH)2)). Neutral temperature–solubility, on the other hand, has no definite variation of solubility when there is a change in temperature (solute B (e.g. NaCl and CaSO4)). The solubility of all solutes can be controlled by controlling the temperature and the supersaturation state is reached, resulting in nucleation. However, the effect of temperature control on the difference of solubility is different as shown in Fig. 1. In the case of solutes A and B, supersaturation can be obtained by the decrease of temperature. According to the decrease of the same temperature, a higher solubility difference can be obtained in solute A. Hence, in the case of solute A, the cooling crystallization technique is suitable to achieve supersaturation. However, the cooling crystallization technique is not suitable for solutes B and C, as their solubilities do not decrease at lower temperatures. Therefore, other techniques such as evaporation, reaction and drowning-out crystallization are more suitable for these solutes (B and C). Evaporation crystallization decreases the solubility of the solute after reaching the supersaturation level once the solvent is evaporated and removed from solution. Drowning-out crystallization induces crystallization after an anti-solvent is added to the solution, thereby decreasing the overall solubility of the solute.

image file: c9ew00157c-f1.tif
Fig. 1 Different types of solubility curves of solutes in the solvent.

Based on the different solubility tendencies at different solution temperatures, temperature control can be used for salt separation from the mixture (solution mining). For example, sylvinite, a sedimentary rock, is a mixture of sylvite and halite, which are KCl and NaCl, respectively. The concepts of crystallization may be exploited to extract KCl and NaCl separately from this mineral. Upon dissolution of sylvanite, K+, Na+, and Cl ions would be present in the solution. Approximately 66% of sylvanite is constituted by NaCl,37 thus Na+ would be present in a greater amount than K+. Upon evaporation of the solution, NaCl (solute B in Fig. 1) precipitates, leaving a solution containing mostly K+ and Cl ions. Following the concepts of crystallization, KCl (solute A in Fig. 1) can then be extracted from the solution upon crystal formation after a marked decrease in temperature. The decrease in temperature will allow supersaturation of KCl in the solution, and the subsequent crystallization to proceed.

Based on the solubility product (Ksp) and ionic product (IP) of a substance, the potential of solute crystallization and precipitation can be predicted:38

• When Ksp > IP, crystallization cannot occur since supersaturation is not reached.

• When Ksp = IP, the solution is saturated, but not enough for crystallization to occur.

• When Ksp < IP, crystallization occurs due to the solution being supersaturated.

As depicted in Fig. 2, there are three zones in accordance with liquid and solid phase variation: stable zone, metastable zone and unstable zone, and a certain amount of energy is needed for phase variation which is from a liquid to a solid. Regardless of the crystallization technique employed, crystal formation and growth occur in the saturation zones including metastable and unstable zones. Spontaneous nucleation occurs in the unstable zone (supersaturation), and crystals can be grown in the metastable zone. However, nucleation does not occur in the metastable zone. On the other hand, both nucleation and crystal growth do not occur in the stable zone. For nucleation, the concentration of salt in a solution should reach the unstable zone, and energy is needed. The energy barrier (energy requirement) for phase variation is dependent on the salt solubility in the solution. It can be controlled by controlling the solubility. The control of solution temperature is the one of the methods for that. As shown in Fig. 2, the solubility is lower at lower temperature for some salt contents, and nucleation can be achieved with a lower amount of energy than that at higher temperature. This means that the supersaturation (unstable zone) at lower concentration of solution is reached with a low energy barrier compared to that at high temperature. Stimulation and control of nucleation by control of solute solubility have been widely reported.21,39–42 All these studies aimed to introduce and establish suitable conditions for nucleation, and they included a reduction of the energy barrier of nucleation, resulting in nucleation.43,44 In addition, the control of the saturation level (metastable and unstable zones) can determine the critical nucleation size, crystal size distribution (CSD) and crystal induction time. Seeding crystallization is the one of the options for the stimulation of crystallization without reaching the supersaturation level (unstable zone). Seeding crystallization is commonly practiced in industrial crystallization processes. The presence of seed in the solution allows crystals to grow in the metastable zone.

image file: c9ew00157c-f2.tif
Fig. 2 Liquid and solid phase variation as a function of temperature.

Current approaches to the crystallization process may be well-established and widely used in industries. There are, however, limitations which influence the quality of crystals produced and overall process efficiency. The main limitations are as follows:3,45 (1) poor reproducibility of the produced crystals in terms of quality and characteristics, due to inconsistent agitation and limited supersaturation level control, (2) reduction of the supersaturation level after crystallization, which makes it difficult to continue operations without additional treatment, and (3) high energy consumption for heating in conventional evaporation crystallization. In this context, a new concept and approach in crystallization using membrane technology is being investigated to overcome the limitations of conventional crystallization technologies.

3. Membrane distillation and crystallization process

3.1 Principles of MD/MDC

The membrane distillation (MD) process is a thermal separation strategy and it has various applications, including the treatment of mine water, wastewater, radioactive waste water, brackish water, seawater and reverse osmosis (RO) brine (concentrate).46,47 It is considered to be one of the most attractive technologies for seawater desalination applications. Normally, MD has some advantages including a lower hydraulic pressure than the RO process, high rejection capacity of non-volatile elements, lower operation temperature and smaller footprint compared to conventional distillation processes.19 In the MD process, the membrane, which is hydrophobic, is placed between a high temperature feed and a low temperature permeate. This hydrophobic property leads to only vapor molecules passing through the membrane (Fig. 3).
image file: c9ew00157c-f3.tif
Fig. 3 Concentration, temperature and vapor pressure profile over the membrane in the direct contact membrane distillation (MD) process (image adapted from previous studies51,52) (Tf is the temperature in the feed solution, Tfm is the temperature on a membrane surface in the feed solution stream, Tp is the temperature in the permeate stream, Tpm is the temperature on the membrane surface in the permeate stream, Cf is the concentration in the feed solution stream, Cfm is the concentration on the membrane surface in the feed solution side, Cp is the concentration in the permeate stream, Pf is the hydraulic pressure of the feed solution, and Pp is the hydraulic pressure of the permeate).

Therefore, in the MD process saline water and wastewater can be converted into high-quality water (permeate side) and a concentrate having the same components as the mother liquid, but at a higher concentration (feed side). The main driving force for such separation is the vapor pressure gradient (ΔP = PfPp) resulting from a temperature difference (ΔT = TfmTpm) between the hot feed (f) and the cold permeate (p).4 The conditions of temperature, concentration and hydraulic pressure in direct contact membrane distillation (DCMD) are illustrated in Fig. 3. The concentration polarization (CP) and temperature polarization (TP) lead to different real concentrations and temperatures at the membrane surface between the bulk solution and permeate,21,48,49 resulting in a loss of driving force and mass transfer.50 The performance of MD is limited by CP/TP phenomena. These resist the heat and mass transfer at the boundary layer near the membrane. Hence, having a precise understanding of these phenomena is very important for appropriate operation of MD/MDC.

In the MD process, there are four basic configurations depending on the scenario of the management of the permeate side (Fig. 4): (a) direct contact membrane distillation (DCMD), (b) vacuum membrane distillation (VMD), (c) air gap membrane distillation (AGMD), and (d) sweep gas membrane distillation (SGMD). Each configuration has its own advantages and disadvantages; however, of the four types,14 DCMD is the most studied MD configuration. It has attracted more than half of all studies on MD due to its simple design and operation. Despite these advantages, DCMD does report the poorest energy efficiency for all the configurations. In the VMD process, the permeate side of the membrane is operated under vacuum conditions or under low pressure. In the VMD configuration, a higher water flux compared to the other configurations can be achieved because of the higher driving force and prevention of boundary layer formation near the membrane.19 However, more severe fouling problems are caused which in turn lead to the process performance becoming degraded.14 The AGMD configuration is considered to have the highest energy efficiency and is popular in commercial applications. The air gap width is usually in the 1–10 mm range, which is much larger than the membrane thickness (usually less than 300 μm). It influences the water flux of MD negatively because a membrane's thickness influences the mass transfer of MD. In SGMD, the gas is sweeping in the permeate side and the vapor is transferred and condensed in an external condenser. It has a lower conduction heat loss than DCMD.

image file: c9ew00157c-f4.tif
Fig. 4 Different configurations of the membrane distillation process: (a) direct contact membrane distillation (DCMD), (b) vacuum membrane distillation (VMD), (c) air gap membrane distillation (AGMD) and (d) sweep gas membrane distillation (SGMD).

One recent innovation known as membrane distillation crystallization (MDC) has been investigated, thanks to the development of new membrane materials and membrane processes. MDC consists of both membrane distillation (MD) and crystallization in which high-quality water is produced in the MD process while the production of crystals can be achieved from the concentrated solution in a crystallizer.4,53 These processes occur at the same stage which can help in the recovery of valuable products and the production of fresh water from seawater desalination and treatment of reverse osmosis (RO) and nanofiltration (NF) brine. In high concentration wastewater treatment using the MDC process, valuable resources such as salt, lithium, and so on can be recovered because the thermally driven operation concentrates an aqueous solution up to the supersaturation state.

Drioli and Wu54 first reported the applicability of the crystallization process in the recovery of chemicals from the feed solution. The MD process can be used to reach supersaturation by further concentration of the solution. Wu et al.55 examined the MD–crystallization process to recover taurine in industrial wastewater treatment. Their findings highlighted good prospects for the MD–crystallization process in the treatment of industrial wastewater. Both high quality water and crystals were obtained in their investigation. In these studies, the cooling crystallization technique was used in the individual crystallizer (Fig. 5(a)). In another study, Gryta51 utilized an integrated MD–crystallization process for the treatment of concentrated sodium chloride solution in both batch and continuous modes. In this study, DCMD was used and furthermore, Gryta51 stated that a sodium chloride (NaCl) crystal of 100 kg m−2 d and a permeate flux of 400 L m−2 d could be obtained when using an integrated MD and crystallization process. As shown in Fig. 5(b), the feed tank was replaced by a crystallizer without applying the cooling crystallization technique (temperature of the crystallizer: 323, 290, and 353 K, respectively).

image file: c9ew00157c-f5.tif
Fig. 5 Experimental set-up for membrane distillation crystallization (MDC), reproduced from (a) ref. 55 with permission from Elsevier, copyright 1991, and (b) ref. 51 with permission from Taylor & Francis Online.

In the membrane crystallization technique, the membrane can be used as the tool and method for generating supersaturation. The membrane's function is different depending on the membrane material and design parameters. For the generation of supersaturation, the membrane can function in terms of heat transfer (heat exchanger), mass and heat transfer (membrane distillation and pervaporation), selective mass transfer (ultrafiltration, nanofiltration, reverse osmosis, and ion exchange) or non-selective mass transfer for the purpose of mixing reactants (membrane contactor (MC)).56 Drioli et al.3 stated that the function of the membrane in a membrane crystallizer is to remove the solvent. They asserted that the membrane in this case does not act as a physical barrier for selective mass transport of specific components. Nonetheless the membrane can support the generation, maintenance and control of a supersaturation environment in which nucleation and growth of crystals occur. The uniform size and controlled morphology of crystals produced can be due to the supersaturation environment created by the membrane. This is because of the homogeneous nucleation of crystals at numerous points (pores of the membrane) in which solvent removal occurs.57

In MDC, which is one example of membrane crystallization technology, the membrane helps in the generation of supersaturation. This is made possible by the simultaneous effect of both the removal of the solvent by evaporation and decline in the feed solution temperature. The solvent is continuously removed via solvent evaporation through the membrane from crystallizing the liquid side to the distillate side, resulting in supersaturation of the crystallizing liquid. Moreover, the crystallizing liquid can be cooled down by the heat transferred from the cool liquid. The solubility of the crystalline salt at a low temperature is lower than that at a higher temperature, and this is referred to as negative-temperature solubility. For this reason, the decrease in temperature by the cool liquid enables the supersaturation condition to be reached. Section 3.2 discusses the status of current research with the relevant technical descriptions.

3.2 Current status

Although reverse osmosis (RO) for desalination and wastewater treatment is considered to be the most efficient process for solving the problem of lack of potable water, certain environmental and economic issues remain.58–61 In particular, the disposal of brine (i.e. RO concentrate) is the main drawback of the RO process. The MDC process has attracted much interest as a new strategy for RO brine treatment.62 Prior studies suggested that the MDC process can contribute to a more integrated desalination strategy for water and mineral recovery from seawater, industrial wastewater and brine. In this respect, the treatment of RO brine using the MDC process has been researched more and more. The number of publications on this topic has increased noticeably since 2010. The advantages of integrated MDC and NF/RO processes are: 1) the reduction in the large amount of disposed brine from NF/RO, 2) improved overall fresh water recovery and water quality, and 3) the recovery of valuable resources from brine, resulting in more economical MDC operation costs.3

In MDC, the persistence of a significant distillate flux can be achieved by the formation of crystals leaving the mother liquor (feed solution) even though it causes a reduction of supersaturation. Ji et al.8 stated that a high total water recovery factor (88–89%) can be obtained in the treatment of seawater brine by MDC. Drioli et al.2 demonstrated that an overall fresh water recovery of up to 92.8% can be obtained during RO brine treatment. They also suggested that the MDC process leads to zero liquid discharge (ZLD) due to high water recovery being achieved. Integrated MDC has a clear advantage in reducing the amount of brine disposal and producing additional crystalline salt. The produced crystals and the increase of water recovery can offset the entire operational costs. Subsequently, the adverse influence of brine on the environment, which is caused by its high concentration and chemical content, can be mitigated or eliminated by the continuous recovery of fresh water and resources in MDC.14,63 Macedonio et al.27 stated that by combining different membrane processes, one can achieve better performance for seawater desalination. A much higher amount of water recovered (92%) can be achieved using a combination of NF, RO and MDC compared to a single RO unit (about 40%) and typical multi-stage flash (MSF) (about 10%).

The representative simultaneous MDC process designs are shown in Fig. 6.15,53,64 The crystallizer placed in the retentate (feed) side of the MD process serves as the feed tank as well. The retentate from the membrane module of MD is transferred to the crystallizer which increases the saturation level in it, resulting in crystallization (Fig. 6(a) and (b)). In these studies, the MDC process behaved as though it was a batch-type evaporation crystallizer. When the state of supersaturation was attained, crystallization occurred in the crystallizer with no further evaporation and the solvent being removed by MD.53 There are different functions of crystallization, which increase the saturation level of the retentate in the MDC process design: (1) the removal of the solvent by evaporation (non-control of temperature) (Fig. 6(a) and (c)) and (2) the decrease of temperature in the crystallizer with the solvent being removed (Fig. 6(b)).

image file: c9ew00157c-f6.tif
Fig. 6 Schematic diagrams of MDC with different membrane module configurations (figures adapted from previous studies): (a) with the DCMD-flat sheet membrane module of Tun et al. (2005) (reproduced from ref. 53 with permission from Elsevier, copyright 2005), (b) with DCMD of Edwie and Chung (2012) (reproduced from ref. 15 with permission from Elsevier, copyright 2012), and (c) with hollow fiber submerged VMD of Julian et al. (2016) (reproduced from ref. 64 with permission from Elsevier, copyright 2016).

However, in the case of the latter, the supersaturation level can be reached faster when the solubility of the target component is influenced by temperature. For example, the solubility of Na2SO4, KCl and MgCl2 is significantly influenced by temperature. Their solubility is lower at a lower temperature of solution than at a high temperature, and this is termed negative temperature–solubility correlation. The application of cooling (decrease of temperature) in a crystallizer (cooling crystallization technique) is a suitable application in this case for stimulating crystallization.62 The DCMD process has been widely used in MDC because it is easy to operate. However, the feasibility of different MD configurations such as vacuum membrane distillation (VMD) and osmotic membrane distillation (OD) in the MDC process has been investigated.1,64–67 Julian et al.64 examined the submerged VMD–crystallization process for inland brine water treatment, and found that the need to reheat the feed solution (which transferred from the crystallizer) and the heat loss due to circulation of the feed solution can be eliminated and avoided altogether by using a submerged membrane module. VMD also has higher mass transfer as compared to cross-flow DCMD. The MDC process has been researched for recovering water and resources in challenging solutions such as SWRO brine, produced water from the oil field (shale gas production), and industrial wastewater. The status of current research on MDC is summarized in Table 1.

Table 1 MDC research on the treatment of challenging solutions
Source Feed solutiona MDC configurationb Crystallization techniquec Operative temperature (°C) Crystallizer locationd Recovered mineral Year Ref.
Feed Permeate Crystallizer
a SWRO = sea water reverse osmosis, RO = reverse osmosis, NF = nanofiltration, R = water recovery factor. b DCMD = direct contact membrane distillation, VMD = vacuum membrane distillation, OMD = osmotic membrane distillation, Cr = crystallization, HF = hollow fiber membrane. c “Evaporative” indicates that additional treatment (such as temperature control or chemical treatment) is not applied; only by concentration by the MD process. “Cooling” indicates that the temperature in the crystallizer was maintained below the temperature of the bulk solution. d This indicates the crystallizer installation location. There are two types: MD and crystallizer separately (“Batch-type”) and combination both MD and crystallizer in one process (“Feed line” and “feed tank (the crystallizer plays the role of the feed tank)”).
Single component solution Synthetic NaCl solution DCMD-Cr (HF) Evaporation & cooling 60.0–85.0 20.0–55.0 20.0 Batch-type or feed line (=feed tank) NaCl 2002 51
Na2SO4 solution (2.0 M) DCMD-Cr (flat sheet) Evaporation 60.0 20.0 70.0 ± 2.0 Feed line (=feed tank) Na2SO4 2005 53
NF brine (raw feed: 60 g L−1 Na2SO4, R = 50%) DCMD-Cr (HF) Evaporation or cooling 30.0 25.0 35.0 Additional continuous line Na2SO4 2010 68
040.0 28.0
NaCl solution (24.0 wt%) DCMD-Cr (HF) Evaporation & cooling 60.0 ± 0.1 17.0 ± 0.1 Feed line (=feed tank) NaCl 2012 15
80.0 ± 0.1
NaCl solution (70.0 g L−1) DCMD-Cr (HF) Evaporation & cooling 60.0–80.0 30.0–50.0 30.0 Feed line (=feed tank) NaCl 2012 9
NaCl solution (26.4 wt%) DCMD-Cr (HF) Evaporation & cooling 40.0 ± 0.1 17.0 ± 0.1 25.0 ± 0.1 Feed line (=feed tank) NaCl 2013 31
70.0 ± 0.1
NaCl solution MD Evaporative 2013 4
CaCO3 solution OD Osmotic
NaCl solution (26.7 wt%) DCMD-Cr (HF) Evaporation & cooling 64.9 29.9 <64.9 Feed line NaCl 2014 29
KNO3 solution DCMD-Cr (HF) 45.0–70.0 20.0 & 30.0 25.0 Feed line (=feed tank) KNO 2015 21
NH4Cl solution (20.0%) VMD-Cr (flat-sheet) Evaporation & cooling 60.0 10.0 Batch type (after MD) NH4Cl 2015 65
NH4SO4 solution (34.0%)
LiCl solution (6.0 M) DCMD-Cr Evaporation & cooling, osmotic 52.0 20.0 30.0–60.0 Feed line (=feed tank) LiCl 2016 1
Solution (7.0 M) OMD-Cr 52.0 20.0
Solution (8.0 M) VMD 40.0–60.0
Na2SO4 solution (0.8 M) Submerged-DCMD-Cr (HF) Evaporation & cooling 50.0 ± 1.3 16.5 ± 0.2 20.0 ± 1.5 Feed line (=feed tank) Na2SO4 2018 62
Brine (from desalination process) Modeled NF brine DCMD-Cr (HF) Evaporation & cooling 35.0 ± 1.5 15.0 ± 1.0 25.0 Feed line (=feed tank) MgSO4·7H2O 2004 69
NF/RO brine DCMD-Cr (HF) Evaporation & cooling 34.0 ± 1.0 16.0 ± 2.0 25.0 Feed line (=feed tank) NaCl 2010 38
SWRO brine DCMD-Cr (HF) Evaporation & cooling 40.0 20.0 30.0 Feed line (=feed tank) NaCl 2010 8
SWRO brine (R = 51%) DCMD-Cr (HF) Evaporation 38.6 ± 0.4 21.3 ± 0.1 38.6 ± 0.4 Feed line (=feed tank) NaCl 2013 70
Seawater SWRO brine (R = 51%) DCMD-Cr (HF) Evaporation 36.1 ± 1.4 24.7 ± 2.0 36.1 ± 1.4 Feed line (=feed tank) MgSO4·7H2O 2016 32
Sea salt solution (65 g L−1) DCMD-Cr (HF) Evaporation 60.0 20.0 60.0 Feed line (=feed tank) NaCl 2016 71
Inland brine Submerged-VMD-Cr (HF) Evaporation & cooling 70.0 ± 0.5 <70.0 ± 0.5 Feed line (=low part of feed tank) 2016 64
Real Concentrated seawater (RO brine, R = 30%) DCMD-Cr (HF) Evaporation & cooling 40.0 20.0 30.0 Feed line (=feed tank) NaCl 2010 8
Produce water Modeled Produced water DCMD-Cr Cooling 38.0 25.0 15.0 2017 6
Shale gas produced water DCMD-Cr (flat-sheet) Evaporation & cooling 60.0 ± 0.5 20.0 ± 0.5 40.0 ± 1.0 Feed line (=feed tank) BaCl2 2016 7
Real Oilfield produced water DCMD-Cr (HF) Evaporation 35.0, 45.0, 55.0 10.0 Feed line (=feed tank) NaCl 2015 5
Shale gas produced water DCMD-Cr (HF) Evaporation & cooling 60.0 ± 1.0 20.0 ± 1.0 30.0 Feed line (=feed tank) CaCO3 2017 33
Wastewater Modeled Radioactive wastewater VMD-Cr (HF) Evaporation & cooling 28.3–35.2 20.0 Feed line (=feed tank) Boric acid 2017 66
Real Wastewater DCMD-Cr (HF) Evaporative & seeding (MgCl2) 45.0–65.0 Feed line (=feed tank) MgNH4PO4·H2O 2017 72
Industrial wastewater (NF brine) DCMD-Cr (HF) Evaporation or Cooling 30.8–51.4 16.5–44.9 37.0 Feed line (=feed tank) Na2SO4 2017 28
Carbon dioxide Na2CO3 solution (150 and 200 g L−1) OD-Cr (HF) Evaporation & concentration (osmotic) Feed line (=feed tank) Na2CO3·10H2O 2013 67

3.2.1 Optimization of operation parameters in MDC. MDC has two key advantages in the treatment of high concentration solutions, and these are: (1) the driving force (temperature gradient) in MD, and (2) continuous extraction of the solute from the solution in the form of crystals, resulting in the maintenance of total contents in the crystallizing solution. In the crystallization phase, a high concentration of solution (supersaturation level) is essential to obtain crystals from the solution. If there is a high concentration of the target material in the feed solution, it is then easier and faster to achieve the supersaturation level. In this context, the researchers have used a single component solution with high concentration as a feed solution in order to examine the feasibility of MDC for the treatment of brine from a desalination plant (NF and RO) (NaCl: ref. 4, 9, 15, 29, 31, and 51, Na2SO4: ref. 53, 64, and 68). Also investigated here was the effect of different operational parameters on the current performance of MDC.

The membrane properties and materials are the critical factors in MDC because the membrane surface is in direct contact with a crystalline solution. Therefore the interaction between the solute and membrane is likely to occur depending on the fluid dynamics, morphology and chemical properties.3 Membranes with increased water flux and low sensitivity to fouling can help achieve stable operation and superior performance. Edwie et al.15 analyzed the effect of membrane material and design on the mass/heat transfer in SWRO brine using three different membranes (single-layer polyvinylidene fluoride (PVDF), dual-layer hydrophobic–hydrophobic PVDF, and dual-layer hydrophobic–hydrophilic PVDF/polyacrylonitrile (PAN) hollow fiber membrane). The authors noted that the prominent membrane characteristics for stable DCMD performance when treating a saturated feed solution were: (1) more compact mixed-matrix morphology under the membrane surface, and (2) a membrane with a smaller pore size along with a macrovoid-free cellular structure.

In order to mitigate the effect of generation and deposition of crystals on the membrane surface, reduce membrane fouling, and stimulate the occurrence of crystallization away from the membrane surface, Choi et al.62 investigated a new concept of the MDC process. This was known as fractional-submerged membrane distillation crystallization (F-SMDC). The submerged DCMD membrane module was placed at the top portion of the reactor, and crystallization occurred at the bottom (lower) portion of the reactor (played the role of the feed tank and crystallizer). This operation is based on the concentration gradient (CG) and temperature gradient (TG) along the reactor height. The higher temperature and lower concentration of the feed solution were maintained at the top portion of the reactor, which is suitable for MD operation while the lower temperature and higher concentration were maintained at the bottom portion of the reactor, which favors crystallization. In the presence of the CG/TG in F-SMDC, higher water recovery (VCF 3.5) and less membrane scaling with a high rate of crystal formation were possible to be obtained.

3.2.2 MDC applications. Brine mining and zero liquid discharge. The majority of previous studies utilized synthetic seawater brine as a feed solution to study the effects of inorganic matter present in NF/RO brine. In this context, Ji et al. conducted a detailed experimental study on SWRO brine to examine the impact of organic matter on crystallization. They observed that the dissolved organic matter influenced the crystallization kinetics, for instance the reduction of magma (mixture of the crystal and solution) density, nucleation and growth ratio of crystals, and permeate flux. The appropriate pre-treatment of seawater brine is essential to mitigate the effects of dissolved organic matter on how well the MDC process performs. In this study more than 90% of water was recovered with a reduction in volumetric wastewater disposal.

Virtually all the elements in the periodic table exist in seawater. It provides the opportunity to recover valuable elements from seawater. Quist-Jensen et al. obtained crystals of lithium chloride (LiCl) from a single component solution using the MDC process. In this study, three MD configurations were used with the crystallization process. In a single DCMD test, a negative flux was observed due to osmotic effects which overcame the thermal effect on the water flux. However, in the combination of DCMD and OD, the driving force was further enhanced by using different concentrations and temperature gradients simultaneously. However, doing so was still not enough to overcome the high solubility of LiCl (15.6 mol kg−1 H2O at 20 °C). When VMD was used, the saturation level of LiCl was reached due to the minimum temperature polarization and reduced resistance to vapor transport within the pores.

Creusen et al.4 studied the integrated MDC process for treating brine. They used a combination of membrane distillation (MD) and osmotic distillation (OD) with internal crystallization. They added NaCl and CaCO3 seeds to prevent crystals from plugging the membrane (Fig. 7). They observed a rapid decline in flux after a saturated NaCl was crystallized. This was due to the deposition of NaCl and CaCO3 on the membrane surface (either with or without NaCl seeds) in MD. In OD, the flux decline was reduced when NaCl and CaCO3 seeds were added. A high water recovery percentage (approximately 99.8%) with crystals being produced was obtained through the integrated MD and OD process.

image file: c9ew00157c-f7.tif
Fig. 7 Proposed integrated MDC process for desalination of seawater (reproduced from ref. 4 with permission from Elsevier, copyright 2013). Produced water treatment. Recently, the application of MDC has been extended to the phenomenon of produced water. The treatment and disposal of produced water (PW) from oil and gas has become a major worldwide issue due to the high demand for oil and natural gas. In oil and gas production processes, a large volume of wastewater is generated. Oilfield wastewater which is known as PW contains various inorganic and organic compounds as well as oil compounds, which can contaminate surface water, groundwater and soil.5,10,73 The treatment of PW was studied using the MDC process.5,6 It was demonstrated that MD/MDC emerged as a feasible technology for the treatment of PW as high quality NaCl crystals of 16.4 kg per cubic meter were separated at a water recovery factor of 37%.

Shale gas is an emerging energy source which can extend the duration of fossil fuel usage by several decades. The shale gas production industry generates a huge amount of produced water. In this scenario, Kim et al.7,33 investigated the feasibility of the MDC (DCMD with a flat-sheet or hollow fiber membrane module) process for the treatment of shale gas produced water. A better stability of water flux and higher water recovery were observed in their study due to the reduction of ionic concentration by forced crystallization in a crystallizer. Calcium and barium influenced MDC performance because these can easily form scalant and foulant crystals on the membrane surface. Also, oil and grease caused a decrease of liquid entry pressure (LEP), resulting in the membrane being damaged. They affected the quality of the water produced and consequently, these components induced membrane wetting which was also triggered by the organic matter present. They suggested that the pretreatment of oil, grease and multivalent ions (calcium and barium) prior to MDC application is essential for the treatment of shale gas produced water.7,33 Furthermore, the optimization of operation conditions such as cross-flow velocity and temperature in the crystallizer can improve the energy efficiency of MDC, specifically improving water recovery and crystal production. Resource recovery from wastewater. The wastewater or wastewater concentrate is also a suitable source for MDC to recover valuable resources since these wastewater streams contain high concentrations of organic and inorganic components. For example, these include ammonia, phosphorus, heavy metals, salts, polycyclic aromatic hydrocarbons and proteins.25,74 You et al.65 investigated the feasibility of the MDC process for treating ammonium-rich wastewater (from fertilizer plants, intensive agriculture, and industrial activities) to recover ammonium salt (NH4Cl and (NH4)2SO4). Recovered ammonium salt from wastewater is advantageous because it is widely used as a fertilizer in agriculture. However, in the absence of ammonium treatment, it leads to environmental contamination when it flows into groundwater.75,76 Quist-Jensen et al.28 studied the treatment of industrial wastewater containing large amounts of Na2SO4 by the MDC process.

Without pre-treatment, MDC produced high quality crystals of Na2SO4 with a narrow crystal size distribution (CSD), constant growth rate and high purity although with high potential of scaling caused by bivalent ions and silica. However, with a pre-concentrated feed solution made possible by NF, low purity Na2SO4 crystals were observed despite the fact that pre-concentration by NF can accelerate the saturation level. This situation was attributed to large concentrations of bivalent ions and silica being incorporated compared to the untreated solution. The potential of implementing MDC in wastewater treatment plants was noted for recovering phosphorus and ammonia in the form of struvite crystals (MgNH4PO4·H2O) from the sludge dewatering process.72 Struvite has a practical application as a fertilizer. In this study, recovered phosphorus (60%) was obtained along with a high water recovery (∼70%), which was 35% higher than that without concentration when utilizing MD.72 The MD and MDC processes made it possible to transport volatile components (such as ammonia) to the permeate stream, resulting in it becoming ammonia-rich. It can directly be used as a fertilizer as well as struvite recovered in the form of crystals.

Jia et al.66 investigated the recovery of boric acid from simulated radioactive wastewater via MDC combined with VMD (VMDC). The wastewater from nuclear power plants contains large amounts of boric acid (about 500 ppm) originating from the primary coolant of the nuclear reactor.77 A concentration factor of around 200 times was achieved with VMDC. In total, 96.0% of nuclides and 99.5% of boric acid rejection were obtained. Furthermore 50% of boric acid was successfully recovered from the feed solution when the temperature was 70 °C. Carbon sequestration. Carbon dioxide (CO2) causes the major greenhouse effect in earth. So, for this reason, it is important to capture and reuse it. The MDC process was used for treatment of carbon dioxide originating from combustion technologies. CO2 is typically captured using absorption with amines and water-based solutions such as sodium hydroxide (NaOH) because it is environmentally friendly. NaOH is used as an alkaline liquid absorbent, and CO2 and hydroxyl ions react as follows:69,78
CO2 + OH → HCO3(1)
HCO3 + OH → CO32−(2)

Carbonate is formed in the solution. The main challenge that has to be solved is recovering the carbonate (as Na2CO3) and agent (NaOH) of alkaline liquid absorbent from the solution which captures CO2. MDC is an efficient technology for recovering carbonate in the form of Na2CO3 from the liquid absorbent. Na2CO3 crystallization can be achieved by osmotic distillation crystallization technology.67,79 The concentration of osmotic solution is a key factor to obtain a viable operation efficiency based on the energy consumption.

3.3 Challenges of MDC

The advantages and feasibility of MDC for treating challenging solutions such as SWRO brine and produced water (the wastewater stream generated in oil and gas industries e.g. shale gas) have been demonstrated in various experimental and theoretical research studies (section 3.2.2). However, there are issues which can seriously undermine the performance of MDC. To enable its stable and efficient operation, certain adverse issues should be solved through additional experimental and theoretical studies. Some of these issues are described in more detail in the following sub-sections.
3.3.1 Membrane scaling. It is worth noting here that the following two types of research use the same framework but have two very different aims: (1) induction of crystallization, and (2) prevention of fouling caused by crystallization. In the MDC process, the membrane plays a role in supporting crystallization, making the preferential nucleation mechanism possible. When crystals are formed near the surface of the membrane, they are transferred to the crystallizer (bulk solution) via hydrodynamic transport (flow of solution). This helps to form the seed of the crystal or nuclei (called the seeding effect), and it results in acceleration of crystallization in the bulk solution.33,71 Homogeneous nucleation in the bulk solution and then the growth of crystals will occur when supersaturation is reached. After that, the prediction of grown crystals leads to reduction of the feed concentration which is caused by the reduction of salt content in the bulk solution. This results in the reduction of concentration polarization in the membrane.

On the other hand, this crystallization phenomenon has an adverse effect on membrane processes. The role of the membrane as a support for crystallization in MDC can lead to the membrane's degradation and a less effective operational performance. Previous studies have widely reported that the water flux decline in MD and MDC was mainly caused by crystal deposition on the membrane surface.26,33,71,80,81 Lee et al.82 reported that there are two explanations regarding how flux is decreased by scale formation in the membrane process. Scale formation by inorganic compounds in membrane systems occurred due to crystallization on the surface (heterogeneous) and in the bulk solution (homogeneous).83 Scale formation is caused by crystallization and hydrodynamic transport mechanisms.33,71,81 Crystallization on the membrane surface blocks it because the deposited scale grows laterally on the surface. This results in the phenomenon of flux decline. In the case of bulk crystallization, the nucleated crystal particles derived from the bulk solution can trigger flux decline because they are deposited on the membrane surface. The fine crystal particles transferred from the bulk solution (crystallizer) cause secondary nucleation on the membrane surface. Fig. 8 illustrates all three scale formation mechanisms in the membrane process. Scale formation is also affected by membrane properties and process conditions.

image file: c9ew00157c-f8.tif
Fig. 8 Schematic diagram of scale formation in the membrane process (reproduced from ref. 82 with permission from Elsevier, copyright 1999).

The prevention of crystal formation (and growth)/deposition on the membrane, which is stimulated by the concentration and temperature polarization (CP/TP), remains a notable challenge in the MDC process.4 As shown in Fig. 3, in the membrane boundary layer (feed solution stream side), the concentration of the feed solution is higher (CP) and the temperature in the feed solution stream is lower (TP) than those of the bulk feed solution. These conditions are favorable for the nucleation and growth of crystals. Moreover, in the MDC process, the concentration of the feed solution is considerably high as the concentration should reach and maintain the supersaturation for nucleation. It also provides a favorable condition for the nucleation and growth of crystals on the membrane surface, which aggravates the membrane fouling caused by crystal deposition.47,84 In addition, the concentration of the feed solution influences the water flux tendency in MD although the impact of the feed solution concentration in the MD process is much less than that in other membrane processes, at least theoretically. If the feed concentration in MD is relatively high, the saturation level is reached much more quickly, resulting in the increase of fouling potential, which is caused by crystals being deposited on the membrane surface (plugging/clogging).9,47 In the MDC process, this phenomenon is more serious compared to a single MD process. However, fouling by crystallization can be controlled by recovering the crystals produced.38 The potential of fouling and the effect of the feed concentration increase as the feed solution becomes saturated in the effort to obtain crystals from the bulk solution.

Moreover, the membrane's topographical and chemical properties influence the formation of crystals on the membrane surface. Its porous structure acts as a heterogeneous nucleation support mechanism. The solute molecules are entrapped on the rough morphology surface of the polymeric membrane. This then leads to the interaction with the chemical functionalities located on the membrane's polymeric chain, resulting in more heterogeneous nucleation. It induces supersaturation of the solute at the boundary layer of the membrane. The concentration polarization in which a high concentration layer is found near the membrane surface causes crystal deposition on the membrane. As a result, the lower energy barrier for heterogeneous nucleation near the membrane surface is induced by the presence of the membrane and concentration polarization with homogeneous nucleation caused in the bulk solution.53,64 Ali et al.5 demonstrated the effect of membrane material and morphology on scaling formation. A cavity-like structure on the outside of the membrane (PP) can cause severe scale formation on the membrane's surface because it acts as an anchoring point for crystals to adhere to and deposit on. The PVDF membrane has a low surface roughness but also higher resistance to scale formation.

The degree of membrane crystallization also depends on the operation conditions and other factors such as the flow rate and solution temperature. Fouling caused by generation/deposition of crystals on the membrane surface can be controlled by optimizing the operational factors.38 Shin and Sohn71 found that the effect of the flow rate on scale formation between surface crystallization and bulk crystallization is ambiguous. Surface crystallization decreased with an increase in the flow rate. Bulk crystallization increased with an increase in the flow rate because of secondary nucleation. Also, the low flow rate of the feed solution creates a greater temperature gradient along the membrane module,15 resulting in crystallization occurring on the membrane. Therefore, it is important for the operations to have an appropriate flow rate to reduce the fouling problem.

Studies conducted using real wastewater and SWRO brine treatment show that the water flux of MD was significantly reduced by inorganic scaling such as calcium-based compounds (deposition on the membrane surface).3,4,14,59 This is difficult to manage and control. This scaling problem still remains as a precursor of fouling as the crystals deposited on the membrane surface act as nuclei, resulting in the easy growth of crystals although the concentration of the feed solution does not reach the supersaturation (unstable zone) (seed effect). Growth of crystals leads to clogging of the membrane surface and pores, resulting in the decline of the water flux and recovery ratio. The deposition and formation of sparingly soluble salt crystals is a severe obstacle in improving the water recovery ratio. In desalination and concentrate treatment processes, crystallization on the membrane surface caused by calcium-based crystalline compounds such as calcium carbonate (CaCO3) and calcium sulfate (CaSO4) occurs early. This is because of their lower solubility when compared to other crystalline compounds (such as NaCl, Na2SO4, MgCl2, MgSO4 and KCl) present in seawater and the concentrate.14,59,64,85,86

The fine crystals generated on the membrane surface cannot be removed easily by membrane surface washing with pure water.14,29 Choi et al.14 and Julian et al.87 used air backwashing as an alternative method for removing crystals that formed on the membrane surface; this strategy is suitable for hydrophobic membranes. They observed that every method cannot remove the crystals existing on a membrane surface. Hence, the appropriate control of the entrapped solute on the membrane is very significant both for: firstly, prevention of membrane fouling caused by crystal deposition, and secondly, enhancing the formation of crystals in the bulk solution.

In order to avoid CaSO4 and CaCO3 precipitation during the concentration of NF/RO brine, a conventional lime and soda-ash treatment can serve as a pre-treatment of the NF/RO desalination process. This method reduces the calcium and magnesium hardness of the RO concentrates and thus limits the scaling problem.8,88,89 Adding Na2CO3 into the feed solution was done to remove Ca2+ ions.32,38 Ca2+ ions have been precipitated as CaCO3 through reactive precipitation with anhydrous sodium carbonate.70 98% of Ca2+ ions have been precipitated when Na2CO3 was added at a molar ratio of Ca2+/CO23− of 1[thin space (1/6-em)]:[thin space (1/6-em)]1.05. Creusen et al.4 noted that the seeding of CaCO3 to some extent mitigated CaCO3 scaling in MD. The results indicated that fouling on the membrane surface caused by crystallization of CaCO3 declined by adding CaCO3 seeds in the MD process. This resulted in a sustainable flux of 6 L m−2 h−1 for a long period of time.

The injection of CO2 gas into the bulk solution can also remove Ca2+ as CaCO3 precipitation.69 In order to remove Ca2+ in the feed solution, a membrane contactor was used with carbon dioxide (CO2) gas injection prior to MDC operation. Reactive precipitation was achieved by the addition of NaHCO3/Na2CO3 aqueous solution into the membrane contactor and NF retentate solution (1[thin space (1/6-em)]:[thin space (1/6-em)]1 molar of Ca2+/CO32− ratio). These steps made it possible to remove Ca2+ as CaCO3 precipitation. For stable operation and a higher recovery factor in MDC, the management of the scaling problem is essential, and it can be achieved by the appropriate pretreatment of the feed solution.

3.3.2 Membrane wetting. In MD operation, fouling on the membrane surface aggravates the phenomenon of membrane wetting.47,51,90,91 Penetration of liquid through the hydrophobic membrane should be avoided because it degrades the quality of produced water and damages the membrane, causing ions to be rejected. Tun et al.53 noted that a difference in high bulk temperature may increase membrane wetting. Also, the feed solution's supersaturation state around the membrane surface causes membrane wetting. A wetted membrane cannot be used further in the MD process for producing freshwater from a feed that has high salinity. Membrane wetting seriously compromises the MD process and the quality of water produced (permeate) is undermined by crystallization on the membrane surface. It is associated with the solubility of substances such as salt and scalants, which depends on the temperature of the bulk solution. Temperature control is thus very important. Moreover, Gryta51 stated that the all membrane pores can be filled by the feed solution under these conditions and the wetting phenomenon occurs (Fig. 9). As a result, diffusive transport of the solute through wetted pores occurs, specifically from the feed solution to the permeate side due to the feed and membrane interaction. This in turn causes the permeate water to be contaminated.
image file: c9ew00157c-f9.tif
Fig. 9 Model of the membrane wetting phenomenon (initial solute: ●, solute after Δt: ○) (reproduced from ref. 51 with permission from Taylor & Francis Online, copyright 2002).

The deposition of organic components on the hydrophobic membrane leads to a decrease in membrane hydrophobicity in the MD/MDC process, resulting in the wetting phenomenon as well as organic fouling.63 Specific components in the feed solution cause the solute to penetrate the membrane pores. For example, in the treatment of shale gas produced water (SGPW), membrane wetting caused by organic, inorganic components and oil matter resulted in a relatively low recovery (from 20–25%) of MDC.7 The authors suggested that pretreatment should thus be adopted to remove organic and oil components prior to the application of MDC. Like the scaling problem, the removal of causative substances in the feed solution is the necessary.

3.3.3 Energy efficiency. In most previous membrane distillation crystallization (MDC) research, cooling crystallization has been used as a method to control the degree of saturation of solution (concentration of the feed solution) because of its convenience. When it is combined with the MD process, the operational costs will increase significantly because of the higher temperature requirement for MD operation than the cooling crystallization process.4,15,29,64 After cooling crystallization, the feed solution should be reheated for MD operation, and the feed solution is cooled again in a cooling crystallizer to increase the degree of saturation of the solute. These two processes are repeated continuously. The consequence is that a large amount of energy is consumed and wasted, resulting in an increase of the total operation cost.

Guan et al.9 evaluated the consumption of energy in the MDC process. The effect of the heat recovery unit was also investigated using the gain output ratio (GOR) which is a factor commonly used to assess energy efficiency in the evaporation process. They noted that the heat requirement in MDC causes high energy consumption,65 which amounted to 97.8% of the total operation energy. Energy consumption in crystallization is less than 0.5%. Moreover, they suggested that optimized operational conditions of feed stream temperature can ensure that energy is consumed more effectively in MDC.

Drioli et al.92 compared the economic efficiency of the desalination process using two desalination hybrid systems, namely, NF/RO and NF/RO/MDC. The thermal energy demand in the latter hybrid system can be offset by improving water recovery (100%), removing disposed brine and recovering valuable resources However, as mentioned above, when the cooling crystallization method is being implemented in the MDC process, it will result in serious economic problems. In order to improve the energy efficiency of MDC, further research on low energy/cost consumption crystallization methods is critical. Research on other adoptive crystallization methods (for example, cooling) is needed in combination with the MD process in order to improve the MDC process. Mechanical vapor recompression (MVR) is a thermal distillation technology that is widely applied in wastewater processes and seawater desalination.93,94 In MVR, the evaporation and compression mechanism is used under thermal conditions. One of the main advantages of MVR is its capacity to reuse the latent heat of the secondary steam produced in the evaporator. Hence, MVR can achieve higher energy efficiency compared to other thermal processes such as multi-stage flash and multi-effect evaporation. Integrating an MD crystallizer with MVR may potentially improve the energy efficacy of MD.94 However, the additional capital cost due to integration with MVR must be taken into consideration, given the complex mechanical structure of MVR.

3.4 Recommended crystallization techniques for MDC

The application of adaptive crystallization techniques is one of the options to address the adverse issues in MDC. The application of the crystallization technique, which does not use an additional heat source for heating and cooling on crystallization, can improve the energy efficiency in MDC compared to conventional MDC with the cooling crystallization technique. In the MDC process, recently, the cooling crystallization technique has been widely studied. However, the combination of MD and cooling crystallization techniques such as solute crystallization, freeze crystallization and eutectic freeze crystallization leads to high energy consumption. The crystallization techniques which require a lower temperature of the feed solution for crystallization are not sustainable for the MDC process because this feed solution needs to be heated again. As a separate MDC process, the cooling crystallization technique functions as two different processes. However, in a simultaneous MDC process, i.e. where the crystallizer is placed along the feed flow channel, there will consequently be additional heat loss due to reheating of the cooled feed solution that was transferred from the crystallizer. The cooling crystallization technique has limitations in recovering the negative temperature–solubility solute because reaching the supersaturation level of the negative temperature–solubility solute in solution cannot be stimulated by the decrease of temperature.

Additionally, the deployment of an appropriate crystallization technique in MDC can mitigate the membrane scaling and wetting issues. Membrane scaling and wetting issues are caused by the nucleation and deposition of crystals on the membrane surface. The crystals transferred from the crystallizer contribute to making these issues serious. If the crystal transfer is prevented and the crystallization occurrence is controlled in only the crystallizer, the negative issues caused by the crystals on the membrane surface can be reduced. In the following two sub-sections, two possible crystallization technologies that can be adopted for solving the challenges in MDC are discussed.

3.4.1 Reaction crystallization. In reaction crystallization (RC), a reaction between a gas and a solution is used to generate sparingly soluble or insoluble components so that the separation of solutes from the solution can be facilitated.95 A reactant is used to reduce the solubility of the crystallizing component in the solution by a chemical reaction.96 This technique is commonly used in the recovery and separation of inorganic components and heavy metal ions.97–100 Heavy metals ions can be removed/precipitated by converting them into hydroxide, sulfide or carbonate precipitants. The common method is to add alkaline chemicals.25 Hydroxide precipitation is widely used because of its simplicity and low cost. However, this technique does experience some drawbacks such as a low precipitation rate and difficulty in treating the lime effluent. The sulphide precipitation method has been employed to remove a wide range of heavy metals, for example Sr4+, Ni2+, Mn2+, Zn2+, Ag+, Cu2+, Pb2+ and Hg2+ and by using various sulfide sources such as Na2S, NaHS, FeS, H2S and CaS.101 The recovery of NH4+ and PO43− can be achieved by struvite reaction crystallization (MgNH4PO4·6H2O) from a wide range of wastewaters such as fertilizer plant wastewater, swine wastewater and leather tanning wastewater.99,102–104 Crystalline struvite can serve as a mineral fertilizer in agriculture.105 Huang et al. reported a novel process for simultaneously recovering NH4+ and PO43− from swine wastewater. They did this by using modified zeolite as an adsorbent and magnesium source for recovering ammonia and nitrogen.

The reaction crystallization discussed above occurs at a relatively high supersaturation level because homogeneous nucleation occurs prior to heterogeneous nucleation. The energy barrier for heterogeneous seeded nucleation is lower compared to homogeneous nucleation. As a result, heterogeneous nucleation occurs at a lower supersaturation level than the homogeneous process.106 In order to reduce the energy required for nucleation in the solution, seed crystals in the supersaturated solution are usually used to initiate crystallization. Here, seed crystals act as the nuclei of crystals.107 Addition of seed crystals into the supersaturated solution leads to their spontaneous growth although the solution concentration only reaches the metastable zone (nucleation does not occur in the metastable zone). This results in reduction of reaction time. Deng et al.108 investigated the induction of crystallization by using fluorapatite/calcite as seed crystals for the removal of fluoride. The same concept has also been applied for the recovery of magnesium, ammonium and phosphate from wastewater.109–111

3.4.2 Drowning-out crystallization. Drowning-out crystallization is the separation of salt from solution by adding an extraneous component, which is a gas, solid, supercritical fluid or liquid.112 Gas or solid drowning-out agents should be soluble in the original solvent, and the liquid drowning-out agent should be miscible with the original solvent.113 It is also important to select the drowning-out agent accordingly as it influences the morphology and structure of the crystal formed.114,115 This component is often referred to as the drowning-out agent, antisolvent, precipitant, salting-out agent, solventing-out agent or watering-out agent. In this process, through the addition of an extraneous component, the solubility of a targeted component will decrease and the supersaturation level will be reached, resulting in crystallization.25 Controlling the level of solubility with the appropriate agents does influence the crystal morphology, structure and yield.

Dammak et al.116 investigated the feasibility of a drowning-out separation process using ethanol to recover polyphenols from olive mill wastewater. The concentration of polyphenol increased considerably after the addition of ethanol to the wastewater, resulting in a highly-concentrated polyphenol isolate (up to 75% (w/w)). The ethanol used as the drowning-out agent was regenerated by batch evaporation. This method is an energy-saving alternative to cooling and evaporation crystallization.117 An energy cost reduction of 63% was obtained as compared to that of three-effect evaporation by the addition of diisopropylamine (DiPA) used as an antisolvent in producing NaCl and Na2CO3.118 In another study, Zijlema et al.119 investigated the recovery of NaCl using antisolvent crystallization with DiPA. The reduction in energy costs (29%) was achieved and favorably comparable to an on-site integrated steam power plant.

On the other hand, the effect of a drowning-out agent on the characteristics of membranes used in the MDC process should be considered because the decrease in liquid entry pressure (LEP) of a membrane degrades its ion rejection ability, resulting in decreased quality of produced water. It strongly depends on the type and concentration of alcohol.120 Ethanol has been widely used as a drowning-out agent. Increasing the ethanol concentration in the solution causes a linear decrease in the membrane's LEP.121

4. Conclusion

Membrane distillation crystallization (MDC) is a promising technology that can simultaneously recover clean water and valuable resources in crystal forms, for treating various challenging hypersaline solutions (seawater reverse osmosis (SWRO) brine, produced water from oil fields and wastewater). It leads to resource recovery and additional water recovery, resulting in near zero liquid discharge theoretically. Previous and ongoing research studies clearly demonstrate the emerging attraction and feasibility of MDC technology for advanced wastewater treatment and desalination with resource recovery.

This review discusses the theoretical background and state of current research on the MDC process in terms of the treatment of these challenging water sources. The challenges related to MDC performance and operations such as crystallization on the membrane surface, high energy consumption, membrane wetting and energy/cost efficiency were discussed. The following conclusions can be made:

• A substantial amount of MDC research has been based on single component solution treatment such as NaCl, Na2SO4 and CaCO3. It has set out to examine the feasibility of the MDC process in high concentration salt solutions and the effect of operational conditions on MDC performance.

• The MDC process can potentially treat challenging solutions (simulated and raw solutions) such as SWRO brine, produced water from oil fields and wastewater, and lead to the recovery of resources from the feed solution.

• There are some adverse issues in the MDC process that need to be ameliorated, such as crystallization on the membrane surface, membrane wetting and high energy consumption. These were aggravated by the high concentration of inorganic contents and specific contents of the feed solution (e.g. oil in produced water; organic matter in wastewater). The appropriate application of pretreatment strategies should be considered to generate stable performance and prevent such problems.

• Cooling crystallization has been widely employed in the MDC process to intentionally induce supersaturation. However, this leads to the consumption of high energy because the feed solution should be reheated when the feed solution is transferred from the cooling crystallizer to the membrane module.

• Applying other crystallization techniques that have no effect on the temperature (i.e. no decrease in the temperature of the feed solution) and adverse phenomena (e.g. crystallization on the membrane surface and membrane wetting) may improve energy and cost considerations in real and practical scenarios. In order to dispel any concerns with the MDC process, further research on energy consumption and adverse phenomena is needed.

• MDC research has been conducted on the lab-scale. In order to apply it in real scenarios, pilot-scale research is necessary. However, the experimental and theoretical information for MDC operation is not enough for scale-up. Hence, further practical research and theoretical analysis should be conducted.

Conflicts of interest

There are no conflicts to declare.


This work was funded by Australian Research Council Discovery Research Grant (DP150101377).


  1. C. A. Quist-Jensen, A. Ali, S. Mondal, F. Macedonio and E. Drioli, A study of membrane distillation and crystallization for lithium recovery from high-concentrated aqueous solutions, J. Membr. Sci., 2016, 505, 167–173 CrossRef CAS.
  2. E. Drioli, E. Curcio, G. Di Profio, F. Macedonio and A. Criscuoli, Integrating Membrane Contactors Technology and Pressure-Driven Membrane Operations for Seawater Desalination: Energy, Exergy and Costs Analysis, Chem. Eng. Res. Des., 2006, 84, 209–220 CrossRef CAS.
  3. E. Drioli, G. Di Profio and E. Curcio, Progress in membrane crystallization, Curr. Opin. Chem. Eng., 2012, 1, 178–182 CrossRef CAS.
  4. R. Creusen, J. van Medevoort, M. Roelands, A. van Renesse van Duivenbode, J. H. Hanemaaijer and R. van Leerdam, Integrated membrane distillation–crystallization: Process design and cost estimations for seawater treatment and fluxes of single salt solutions, Desalination, 2013, 323, 8–16 CrossRef CAS.
  5. A. Ali, C. A. Quist-Jensen, F. Macedonio and E. Drioli, Application of Membrane Crystallization for Minerals' recovery from produced water, Membranes, 2015, 5, 772–792 CrossRef CAS PubMed.
  6. A. Ali, C. A. Quist-Jensen, E. Drioli and F. Macedonio, Evaluation of integrated microfiltration and membrane distillation/crystallization processes for produced water treatment, Desalination, 2017, 434, 161–168 CrossRef.
  7. J. Kim, H. Kwon, S. Lee, S. Lee and S. Hong, Membrane distillation (MD) integrated with crystallization (MDC) for shale gas produced water (SGPW) treatment, Desalination, 2017, 403, 172–178 CrossRef CAS.
  8. X. Ji, E. Curcio, S. Al Obaidani, G. Di Profio, E. Fontananova and E. Drioli, Membrane distillation-crystallization of seawater reverse osmosis brines, Sep. Purif. Technol., 2010, 71, 76–82 CrossRef CAS.
  9. G. Guan, R. Wang, F. Wicaksana, X. Yang and A. G. Fane, Analysis of Membrane Distillation Crystallization System for High Salinity Brine Treatment with Zero Discharge Using Aspen Flowsheet Simulation, Ind. Eng. Chem. Res., 2012, 51, 13405–13413 CrossRef CAS.
  10. A. Fakhru'l-Razi, A. Pendashteh, L. C. Abdullah, D. R. A. Biak, S. S. Madaeni and Z. Z. Abidin, Review of technologies for oil and gas produced water treatment, J. Hazard. Mater., 2009, 170, 530–551 CrossRef PubMed.
  11. M. Çakmakce, N. Kayaalp and I. Koyuncu, Desalination of produced water from oil production fields by membrane processes, Desalination, 2008, 222, 176–186 CrossRef.
  12. R. Bouchrit, A. Boubakri, A. Hafiane and S. A.-T. Bouguecha, Direct contact membrane distillation: Capability to treat hyper-saline solution, Desalination, 2015, 376, 117–129 CrossRef CAS.
  13. J. A. Bush, J. Vanneste and T. Y. Cath, Membrane distillation for concentration of hypersaline brines from the Great Salt Lake: Effects of scaling and fouling on performance, efficiency, and salt rejection, Sep. Purif. Technol., 2016, 170, 78–91 CrossRef CAS.
  14. Y. Choi, G. Naidu, S. Jeong, S. Vigneswaran, S. Lee, R. Wang and A. G. Fane, Experimental comparison of submerged membrane distillation configurations for concentrated brine treatment, Desalination, 2017, 420, 54–62 CrossRef CAS.
  15. F. Edwie and T.-S. Chung, Development of hollow fiber membranes for water and salt recovery from highly concentrated brine via direct contact membrane distillation and crystallization, J. Membr. Sci., 2012, 421–422, 111–123 CrossRef CAS.
  16. M. Gryta, The study of performance of polyethylene chlorinetrifluoroethylene membranes used for brine desalination by membrane distillation, Desalination, 2016, 398, 52–63 CrossRef CAS.
  17. B. Li and K. K. Sirkar, Novel membrane and device for vacuum membrane distillation-based desalination process, J. Membr. Sci., 2005, 257, 60–75 CrossRef CAS.
  18. S. Meng, Y.-C. Hsu, Y. Ye and V. Chen, Submerged membrane distillation for inland desalination applications, Desalination, 2015, 361, 72–80 CrossRef CAS.
  19. K. W. Lawson and D. R. Lloyd, Membrane distillation, J. Membr. Sci., 1997, 124, 1–25 CrossRef CAS.
  20. S. Zuoliang, Y. Qiuxiang and C. Jianxin, Industrial Crystallization: Trends and Challenges, Chem. Eng. Technol., 2013, 36, 1286–1286 CrossRef.
  21. X. Jiang, D. Lu, W. Xiao, X. Ruan, J. Fang and G. He, Membrane assisted cooling crystallization: Process model, nucleation, metastable zone, and crystal size distribution, AIChE J., 2016, 62, 829–841 CrossRef CAS.
  22. J. Hash and O. C. Okorafor, Crystal size distribution (CSD) of batch salting-out crystallization process for sodium sulfate, Chem. Eng. Process.: Process Intesif., 2008, 47, 622–632 CrossRef CAS.
  23. J. Holaň, L. Ridvan, P. Billot and F. Štěpánek, Design of co-crystallization processes with regard to particle size distribution, Chem. Eng. Sci., 2015, 128, 36–43 CrossRef.
  24. M. R. Abu Bakar, Z. K. Nagy, A. N. Saleemi and C. D. Rielly, The Impact of Direct Nucleation Control on Crystal Size Distribution in Pharmaceutical Crystallization Processes, Cryst. Growth Des., 2009, 9, 1378–1384 CrossRef CAS.
  25. H. Lu, J. Wang, T. Wang, N. Wang, Y. Bao and H. Hao, Crystallization techniques in wastewater treatment: An overview of applications, Chemosphere, 2017, 173, 474–484 CrossRef CAS PubMed.
  26. E. Curcio and E. Drioli, Membrane Distillation and Related Operations—A Review, Sep. Purif. Rev., 2005, 34, 35–86 CrossRef CAS.
  27. F. Macedonio, E. Curcio and E. Drioli, Integrated membrane systems for seawater desalination: energetic and exergetic analysis, economic evaluation, experimental study, Desalination, 2007, 203, 260–276 CrossRef CAS.
  28. C. A. Quist-Jensen, F. Macedonio, D. Horbez and E. Drioli, Reclamation of sodium sulfate from industrial wastewater by using membrane distillation and membrane crystallization, Desalination, 2017, 401, 112–119 CrossRef CAS.
  29. G. Chen, Y. Lu, W. B. Krantz, R. Wang and A. G. Fane, Optimization of operating conditions for a continuous membrane distillation crystallization process with zero salty water discharge, J. Membr. Sci., 2014, 450, 1–11 CrossRef CAS.
  30. R. J. M. Creusen, J. van Medevoort, C. P. M. Roelands and J. A. D. v. R. v. Duivenbode, Brine Treatment by a Membrane Distillation-crystallization (MDC) Process, Procedia Eng., 2012, 44, 1756–1759 CrossRef.
  31. F. Edwie and T.-S. Chung, Development of simultaneous membrane distillation–crystallization (SMDC) technology for treatment of saturated brine, Chem. Eng. Sci., 2013, 98, 160–172 CrossRef CAS.
  32. C. A. Quist-Jensen, F. Macedonio and E. Drioli, Membrane crystallization for salts recovery from brine—an experimental and theoretical analysis, Desalin. Water Treat., 2016, 57, 7593–7603 CrossRef CAS.
  33. J. Kim, J. Kim and S. Hong, Recovery of water and minerals from shale gas produced water by membrane distillation crystallization, Water Res., 2017, 129, 447–459 CrossRef PubMed.
  34. J. W. Mullin, Crystallization, Butterworth-Heinemann, Oxford, 4th edn, 2001 Search PubMed.
  35. H.-H. Tung, E. L. Paul, M. Midler and J. A. McCauley, Crystallization of organic compounds: an industrial perspective, John Wiley & Sons, 2009 Search PubMed.
  36. G. D. Profio, A. Caridi, R. Caliandro, A. Guagliardi, E. Curcio and E. Drioli, Fine Dosage of Antisolvent in the Crystallization of l-Histidine: Effect on Polymorphism, Cryst. Growth Des., 2010, 10, 449–455 CrossRef.
  37. D. E. Garrett, in Potash: Deposits, Processing, Properties and Uses, ed. D. E. Garrett, Springer, Netherlands, Dordrecht, 1996, pp. 325–402,  DOI:10.1007/978-94-009-1545-9_5.
  38. F. Macedonio and E. Drioli, Hydrophobic membranes for salts recovery from desalination plants, Desalin. Water Treat., 2010, 18, 224–234 CrossRef CAS.
  39. O. Narducci and A. G. Jones, Seeding in Situ the Cooling Crystallization of Adipic Acid using Ultrasound, Cryst. Growth Des., 2012, 12, 1727–1735 CrossRef CAS.
  40. R. Lakerveld, N. G. Verzijden, H. Kramer, P. Jansens and J. Grievink, Application of ultrasound for start-up of evaporative batch crystallization of ammonium sulfate in a 75-L crystallizer, AIChE J., 2011, 57, 3367–3377 CrossRef CAS.
  41. N. T. N. Phuong and K. Kwang-Joo, Transformation of hemipentahydrate to monohydrate of risedronate monosodium by seed crystallization in solution, AIChE J., 2011, 57, 3385–3394 CrossRef.
  42. A. Soare, R. Dijkink, M. R. Pascual, C. Sun, P. W. Cains, D. Lohse, A. I. Stankiewicz and H. J. M. Kramer, Crystal Nucleation by Laser-Induced Cavitation, Cryst. Growth Des., 2011, 11, 2311–2316 CrossRef CAS.
  43. J. Ulrich and P. Frohberg, Problems, potentials and future of industrial crystallization, Front. Chem. Sci. Eng., 2013, 7, 1–8 CrossRef.
  44. J. Ulrich and M. J. Jones, Industrial Crystallization: Developments in Research and Technology, Chem. Eng. Res. Des., 2004, 82, 1567–1570 CrossRef CAS.
  45. N. S. Tavare, Micromixing limits in an MSMPR crystallizer, Chem. Eng. Technol., 1989, 12, 1–11 CrossRef CAS.
  46. L. D. Tijing, Y. C. Woo, J.-S. Choi, S. Lee, S.-H. Kim and H. K. Shon, Fouling and its control in membrane distillation—A review, J. Membr. Sci., 2015, 475, 215–244 CrossRef CAS.
  47. G. Naidu, S. Jeong, S. Vigneswaran, T.-M. Hwang, Y.-J. Choi and S.-H. Kim, A review on fouling of membrane distillation, Desalin. Water Treat., 2016, 57, 10052–10076 CrossRef.
  48. S. Srisurichan, R. Jiraratananon and A. G. Fane, Mass transfer mechanisms and transport resistances in direct contact membrane distillation process, J. Membr. Sci., 2006, 277, 186–194 CrossRef CAS.
  49. M. N. Chernyshov, G. W. Meindersma and A. B. de Haan, Modelling temperature and salt concentration distribution in membrane distillation feed channel, Desalination, 2003, 157, 315–324 CrossRef CAS.
  50. M. S. El-Bourawi, Z. Ding, R. Ma and M. Khayet, A framework for better understanding membrane distillation separation process, J. Membr. Sci., 2006, 285, 4–29 CrossRef CAS.
  51. M. Gryta, Concentration of NaCl solution by membrane distillation integrated with crystallization, Sep. Sci. Technol., 2002, 37, 3535–3558 CrossRef CAS.
  52. G. Guan, X. Yang, R. Wang, R. Field and A. G. Fane, Evaluation of hollow fiber-based direct contact and vacuum membrane distillation systems using aspen process simulation, J. Membr. Sci., 2014, 464, 127–139 CrossRef CAS.
  53. C. M. Tun, A. G. Fane, J. T. Matheickal and R. Sheikholeslami, Membrane distillation crystallization of concentrated salts—flux and crystal formation, J. Membr. Sci., 2005, 257, 144–155 CrossRef CAS.
  54. E. Drioli, Y. Wu and V. Calabro, Membrane distillataion in the treatment of aqueous solutions, J. Membr. Sci., 1987, 33, 277–284 CrossRef CAS.
  55. Y. Wu, Y. Kong, J. Liu, J. Zhang and J. Xu, An experimental study on membrane distillation-crystallization for treating waste water in taurine production, Desalination, 1991, 80, 235–242 CrossRef CAS.
  56. E. Chabanon, D. Mangin and C. Charcosset, Membranes and crystallization processes: State of the art and prospects, J. Membr. Sci., 2016, 509, 57–67 CrossRef CAS.
  57. G. Di Profio, E. Curcio and E. Drioli, Trypsin crystallization by membrane-based techniques, J. Struct. Biol., 2005, 150, 41–49 CrossRef CAS PubMed.
  58. J. Morillo, J. Usero, D. Rosado, H. El Bakouri, A. Riaza and F.-J. Bernaola, Comparative study of brine management technologies for desalination plants, Desalination, 2014, 336, 32–49 CrossRef CAS.
  59. Y. Choi, G. Naidu, S. Jeong, S. Lee and S. Vigneswaran, Effect of chemical and physical factors on the crystallization of calcium sulfate in seawater reverse osmosis brine, Desalination, 2018, 426, 78–87 CrossRef CAS.
  60. D. A. Roberts, E. L. Johnston and N. A. Knott, Impacts of desalination plant discharges on the marine environment: A critical review of published studies, Water Res., 2010, 44, 5117–5128 CrossRef CAS PubMed.
  61. J. S. Ho, Z. Ma, J. Qin, S. H. Sim and C.-S. Toh, Inline coagulation–ultrafiltration as the pretreatment for reverse osmosis brine treatment and recovery, Desalination, 2015, 365, 242–249 CrossRef CAS.
  62. Y. Choi, G. Naidu, S. Jeong, S. Lee and S. Vigneswaran, Fractional-submerged membrane distillation crystallizer (F-SMDC) for treatment of high salinity solution, Desalination, 2018, 440, 59–67 CrossRef CAS.
  63. G. Naidu, S. Jeong, Y. Choi and S. Vigneswaran, Membrane distillation for wastewater reverse osmosis concentrate treatment with water reuse potential, J. Membr. Sci., 2017, 524, 565–575 CrossRef CAS.
  64. H. Julian, S. Meng, H. Li, Y. Ye and V. Chen, Effect of operation parameters on the mass transfer and fouling in submerged vacuum membrane distillation crystallization (VMDC) for inland brine water treatment, J. Membr. Sci., 2016, 520, 679–692 CrossRef CAS.
  65. W.-T. You, Z.-L. Xu, Z.-Q. Dong and M. Zhang, Vacuum membrane distillation–crystallization process of high ammonium salt solutions, Desalin. Water Treat., 2015, 55, 368–380 CrossRef CAS.
  66. F. Jia, J. Li and J. Wang, Recovery of boric acid from the simulated radioactive wastewater by vacuum membrane distillation crystallization, Ann. Nucl. Energy, 2017, 110, 1148–1155 CrossRef CAS.
  67. P. Luis, D. Van Aubel and B. Van der Bruggen, Technical viability and exergy analysis of membrane crystallization: Closing the loop of CO2 sequestration, Int. J. Greenhouse Gas Control, 2013, 12, 450–459 CrossRef CAS.
  68. E. Curcio, X. Ji, A. M. Quazi, S. Barghi, G. Di Profio, E. Fontananova, T. Macleod and E. Drioli, Hybrid nanofiltration–membrane crystallization system for the treatment of sulfate wastes, J. Membr. Sci., 2010, 360, 493–498 CrossRef CAS.
  69. E. Drioli, E. Curcio, A. Criscuoli and G. D. Profio, Integrated system for recovery of CaCO3, NaCl and MgSO4·7H2O from nanofiltration retentate, J. Membr. Sci., 2004, 239, 27–38 CrossRef CAS.
  70. F. Macedonio, C. A. Quist-Jensen, O. Al-Harbi, H. Alromaih, S. A. Al-Jlil, F. Al Shabouna and E. Drioli, Thermodynamic modeling of brine and its use in membrane crystallizer, Desalination, 2013, 323, 83–92 CrossRef CAS.
  71. Y. Shin and J. Sohn, Mechanisms for scale formation in simultaneous membrane distillation crystallization: Effect of flow rate, J. Ind. Eng. Chem., 2016, 35, 318–324 CrossRef CAS.
  72. C. A. Quist-Jensen, J. M. Sørensen, A. Svenstrup, L. Scarpa, T. S. Carlsen, H. C. Jensen, L. Wybrandt and M. L. Christensen, Membrane crystallization for phosphorus recovery and ammonia stripping from reject water from sludge dewatering process, Desalination, 2017, 440, 150–160 Search PubMed.
  73. Z. Zhang, X. Du, K. H. Carlson, C. A. Robbins and T. Tong, Effective treatment of shale oil and gas produced water by membrane distillation coupled with precipitative softening and walnut shell filtration, Desalination, 2019, 454, 82–90 CrossRef CAS.
  74. W. Wang, H. Han, M. Yuan, H. Li, F. Fang and K. Wang, Treatment of coal gasification wastewater by a two-continuous UASB system with step-feed for COD and phenols removal, Bioresour. Technol., 2011, 102, 5454–5460 CrossRef CAS PubMed.
  75. S. W. Effler, C. M. Brooks, M. T. Auer and S. M. Doerr, Free Ammonia and Toxicity Criteria in a Polluted Urban Lake, Res. J. Water Pollut. Control Fed., 1990, 62, 771–779 CAS.
  76. C. Y. Li, W. G. Li, G. Z. Wang and K. Wang, The Preparation and the Research of Copper-Chelex Chitosan for Removal Ammonia-Nitrogen from the Drinking Water, Adv. Mater. Res., 2010, 113–116, 1166–1169 CAS.
  77. X. Wen, F. Li and X. Zhao, Removal of nuclides and boron from highly saline radioactive wastewater by direct contact membrane distillation, Desalination, 2016, 394, 101–107 CrossRef CAS.
  78. A. Mansourizadeh, A. F. Ismail and T. Matsuura, Effect of operating conditions on the physical and chemical CO2 absorption through the PVDF hollow fiber membrane contactor, J. Membr. Sci., 2010, 353, 192–200 CrossRef CAS.
  79. I. Ruiz Salmón, R. Janssens and P. Luis, Mass and heat transfer study in osmotic membrane distillation-crystallization for CO2 valorization as sodium carbonate, Sep. Purif. Technol., 2017, 176, 173–183 CrossRef.
  80. S. Kimura, S.-I. Nakao and S.-I. Shimatani, Transport phenomena in membrane distillation, J. Membr. Sci., 1987, 33, 285–298 CrossRef CAS.
  81. O. R. Lokare, S. Tavakkoli, S. Wadekar, V. Khanna and R. D. Vidic, Fouling in direct contact membrane distillation of produced water from unconventional gas extraction, J. Membr. Sci., 2017, 524, 493–501 CrossRef CAS.
  82. S. Lee, J. Kim and C.-H. Lee, Analysis of CaSO4 scale formation mechanism in various nanofiltration modules, J. Membr. Sci., 1999, 163, 63–74 CrossRef CAS.
  83. H.-J. Oh, Y.-K. Choung, S. Lee, J.-S. Choi, T.-M. Hwang and J. H. Kim, Scale formation in reverse osmosis desalination: model development, Desalination, 2009, 238, 333–346 CrossRef CAS.
  84. J. A. Sanmartino, M. Khayet, M. C. García-Payo, H. El Bakouri and A. Riaza, Desalination and concentration of saline aqueous solutions up to supersaturation by air gap membrane distillation and crystallization fouling, Desalination, 2016, 393, 39–51 CrossRef CAS.
  85. G. Naidu, S. Jeong and S. Vigneswaran, Influence of feed/permeate velocity on scaling development in a direct contact membrane distillation, Sep. Purif. Technol., 2014, 125, 291–300 CrossRef CAS.
  86. S. Lee, J.-S. Choi and C.-H. Lee, Behaviors of dissolved organic matter in membrane desalination, Desalination, 2009, 238, 109–116 CrossRef CAS.
  87. H. Julian, Y. Ye, H. Li and V. Chen, Scaling mitigation in submerged vacuum membrane distillation and crystallization (VMDC) with periodic air-backwash, J. Membr. Sci., 2018, 547, 19–33 CrossRef CAS.
  88. F. Mohammadesmaeili, M. K. Badr, M. Abbaszadegan and P. Fox, Byproduct Recovery from Reclaimed Water Reverse Osmosis Concentrate Using Lime and Soda-Ash Treatment, Water Environ. Res., 2010, 82, 342–350 CrossRef CAS.
  89. V. Masindi, M. S. Osman and A. M. Abu-Mahfouz, Integrated treatment of acid mine drainage using BOF slag, lime/soda ash and reverse osmosis (RO): Implication for the production of drinking water, Desalination, 2017, 424, 45–52 CrossRef CAS.
  90. M. Gryta and M. Barancewicz, Influence of morphology of PVDF capillary membranes on the performance of direct contact membrane distillation, J. Membr. Sci., 2010, 358, 158–167 CrossRef CAS.
  91. M. Gryta, Long-term performance of membrane distillation process, J. Membr. Sci., 2005, 265, 153–159 CrossRef CAS.
  92. E. Drioli, A. Criscuoli and E. Curcio, Integrated membrane operations for seawater desalination, Desalination, 2002, 147, 77–81 CrossRef CAS.
  93. Y. Zhou, C. Shi and G. Dong, Analysis of a mechanical vapor recompression wastewater distillation system, Desalination, 2014, 353, 91–97 CrossRef CAS.
  94. Z. Si, D. Han, Y. Song, J. Chen, L. Luo and R. Li, Experimental investigation on a combined system of vacuum membrane distillation and mechanical vapor recompression, Chem. Eng. Process., 2019, 139, 172–182 CrossRef CAS.
  95. H. Huang, D. Xiao, R. Pang, C. Han and L. Ding, Simultaneous removal of nutrients from simulated swine wastewater by adsorption of modified zeolite combined with struvite crystallization, Chem. Eng. J., 2014, 256, 431–438 CrossRef CAS.
  96. B. L. Å. Slund and Å. K. C. Rasmuson, Semibatch reaction crystallization of benzoic acid, AIChE J., 1992, 38, 328–342 CrossRef.
  97. J. Rubio and F. Tessele, Removal of heavy metal ions by adsorptive particulate flotation, Miner. Eng., 1997, 10, 671–679 CrossRef CAS.
  98. K. Suzuki, Y. Tanaka, T. Osada and M. Waki, Removal of phosphate, magnesium and calcium from swine wastewater through crystallization enhanced by aeration, Water Res., 2002, 36, 2991–2998 CrossRef CAS PubMed.
  99. M. M. Rahman, M. A. M. Salleh, U. Rashid, A. Ahsan, M. M. Hossain and C. S. Ra, Production of slow release crystal fertilizer from wastewaters through struvite crystallization – A review, Arabian J. Chem., 2014, 7, 139–155 CrossRef CAS.
  100. S. Tait, W. P. Clarke, J. Keller and D. J. Batstone, Removal of sulfate from high-strength wastewater by crystallisation, Water Res., 2009, 43, 762–772 CrossRef CAS PubMed.
  101. A. H. M. Veeken, L. Akoto, L. W. Hulshoff Pol and J. Weijma, Control of the sulfide (S2−) concentration for optimal zinc removal by sulfide precipitation in a continuously stirred tank reactor, Water Res., 2003, 37, 3709–3717 CrossRef CAS.
  102. R. Yu, J. Geng, H. Ren, Y. Wang and K. Xu, Struvite pyrolysate recycling combined with dry pyrolysis for ammonium removal from wastewater, Bioresour. Technol., 2013, 132, 154–159 CrossRef CAS PubMed.
  103. M. M. Rahman, Y. Liu, J.-H. Kwag and C. Ra, Recovery of struvite from animal wastewater and its nutrient leaching loss in soil, J. Hazard. Mater., 2011, 186, 2026–2030 CrossRef CAS PubMed.
  104. O. Tünay, I. Kabdasli, D. Orhon and S. Kolçak, Ammonia removal by magnesium ammonium phosphate precipitation in industrial wastewaters, Water Sci. Technol., 1997, 36, 225–228 CrossRef.
  105. N. Hutnik, A. Kozik, A. Mazienczuk, K. Piotrowski, B. Wierzbowska and A. Matynia, Phosphates (V) recovery from phosphorus mineral fertilizers industry wastewater by continuous struvite reaction crystallization process, Water Res., 2013, 47, 3635–3643 CrossRef CAS PubMed.
  106. J. T. M. Sluys, D. Verdoes and J. H. Hanemaaijer, Water treatment in a Membrane-Assisted Crystallizer (MAC), Desalination, 1996, 104, 135–139 CrossRef CAS.
  107. K. Wendt, S. Petersen and J. Ulrich, Influence of seeding on concentration distribution within pastilles drop formed out of binary melts, Chem. Eng. Sci., 2015, 133, 70–74 CrossRef CAS.
  108. L. Deng, Y. Liu, T. Huang and T. Sun, Fluoride removal by induced crystallization using fluorapatite/calcite seed crystals, Chem. Eng. J., 2016, 287, 83–91 CrossRef CAS.
  109. T. Zhang, L. Ding, H. Ren and X. Xiong, Ammonium nitrogen removal from coking wastewater by chemical precipitation recycle technology, Water Res., 2009, 43, 5209–5215 CrossRef CAS.
  110. Q. Wu and P. L. Bishop, Enhancing struvite crystallization from anaerobic supernatant, J. Environ. Eng. Sci., 2004, 3, 21–29 CrossRef CAS.
  111. K. N. Ohlinger, T. M. Young and E. D. Schroeder, Postdigestion Struvite Precipitation Using a Fluidized Bed Reactor, J. Environ. Eng., 2000, 126, 361–368 CrossRef CAS.
  112. D. A. Berry, S. R. Dye and K. M. Ng, Synthesis of drowning-out crystallization-based separations, AIChE J., 1997, 43, 91–103 CrossRef CAS.
  113. D. P. Gianluca, S. Carmen, C. Antonella, C. Efrem and D. Enrico, Antisolvent membrane crystallization of pharmaceutical compounds, J. Pharm. Sci., 2009, 98, 4902–4913 CrossRef PubMed.
  114. X. Holmbäck and Å. C. Rasmuson, Size and morphology of benzoic acid crystals produced by drowning-out crystallisation, J. Cryst. Growth, 1999, 198–199, 780–788 CrossRef.
  115. P. A. Barata and M. L. Serrano, Salting-out precipitation of potassium dihydrogen phosphate (KDP): IV. Characterisation of the final product, J. Cryst. Growth, 1998, 194, 109–118 CrossRef CAS.
  116. I. Dammak, M. Neves, H. Isoda, S. Sayadi and M. Nakajima, Recovery of polyphenols from olive mill wastewater using drowning-out crystallization based separation process, Innovative Food Sci. Emerging Technol., 2016, 34, 326–335 CrossRef CAS.
  117. S. Mostafa Nowee, A. Abbas and J. A. Romagnoli, Antisolvent crystallization: Model identification, experimental validation and dynamic simulation, Chem. Eng. Sci., 2008, 63, 5457–5467 CrossRef CAS.
  118. D. A. Weingaertner, S. Lynn and D. N. Hanson, Extractive crystallization of salts from concentrated aqueous solution, Ind. Eng. Chem. Res., 1991, 30, 490–501 CrossRef CAS.
  119. T. G. Zijlema, R. M. Geertman, G.-J. Witkamp, G. M. van Rosmalen and J. de Graauw, Antisolvent Crystallization as an Alternative to Evaporative Crystallization for the Production of Sodium Chloride, Ind. Eng. Chem. Res., 2000, 39, 1330–1337 CrossRef CAS.
  120. M. C. García-Payo, M. A. Izquierdo-Gil and C. Fernández-Pineda, Wetting Study of Hydrophobic Membranes via Liquid Entry Pressure Measurements with Aqueous Alcohol Solutions, J. Colloid Interface Sci., 2000, 230, 420–431 CrossRef PubMed.
  121. C. Gostoli and G. C. Sarti, Separation of liquid mixtures by membrane distillation, J. Membr. Sci., 1989, 41, 211–224 CrossRef CAS.

This journal is © The Royal Society of Chemistry 2019